Development of a Soil Test for Prediction of Long-Term Potential for Phosphorus Loss in Typical Canadian Soils.

The dynamic nature of phosphorus (P) in soils can result in its continuous conversion into forms that are losable to water resources. Risk assessment of soil P must be able to account for long-term potentials of soil P losses. Present environmental soil P tests provide useful information on potentials for immediate losses of soil P, but may not be sufficient for or suitable to the prediction of long-term P release. The objective of this study was to develop a soil P test for predicting long-term soil P loss potential in typical Canadian agricultural soils. Soils of representative types were collected from four typical agro-ecological areas across Canada. Soil P was analyzed using Mehlich-3, Olsen, FeO-strips, and various concentrations of NaOH with and without EDTA at different shaking time periods. The soil P extractable with each extractant was correlated to the total releasable P, TRP, determined with repeatedly P removal for 92 days by resin membrane strip. Soil extractable P determined using all of the methods in the study was highly related to TRP. However, the amounts of soil P extractable with NaOH, both with and without EDTA, were highly closer to TRP than those extractable with Olsen, Mehlich-3 and FeO-strips methods.  Amongst the NaOH extractions, soil P extractable with the combination of 0.025M NaOH with EDTA and 1.0 hour shaking presented the strongest linear relationship to TRP with easy operative implementation and thus can be used to predict long-term P loss potential of agricultural soils.